The inherently opaque nature of soil has made the study of belowground biological processes extremely difficult. Traditional methods include physical excavation and sieving of soil as well as the imposition of artificial viewing surfaces such as are used with rhizotrons and minirhizotrons. These techniques either destroy the geometry of the roots within the soil, or create artifacts by forcing roots to grow along a glass or plastic surface. Many soils of the SE United States supporting pine plantations are light sandy loams, similar in texture to the sand samples used in previous MR imaging experiments. Development of MR methods to identify functional, living roots within soil cores would provide a powerful new tool for the study of belowground processes. The orientation of roots relative to the X-Y plane is especially important, as most nutrient mineralization occurs within the top 10 cm of soil. In addition, description of branching patterns will allow study of the relationship between root architecture and nutrient availability. Understanding of the relationship between belowground C allocation through root production and resource acquisition by roots is currently a topic of great interest in evaluating plant response to global change, and the potential for forests to provide a sink for predicted elevated levels of CO2. I propose to develop imaging protocols, using traditional 3-Dspin echo techniques, to view living roots within intact field soil cores. A limited number of cores will be sampled from the U.S.D.A. Forest Service site in Scotland County, and imaged. Cores will then be dissected, and roots within the core identified. Condition of the recovered root segments will be recorded, and correlated with the ability to detect them within the MR image sets. Core diameter will be 8 cm, well within the size range for current image acquisition protocols.